Zoltán Szadai

Main text (Neuron , ◼◼◼–◼◼◼; May 20, 2026) The authors regret that errors appeared in the originally published manuscript. Figure S1G is correct; however, the corresponding text in the section entitled “3D-RTMC steps” contained errors. Specifically, we used four AOD cards rather than three, and the PMT cards computed the weighted and unweighted sums for the three axes, which were then transferred to and further processed in parallel by the four AOD cards, as shown in detail in Figure S1G.

Keywords two-photon voltage imaging dendritic computation backpropagating action potentials 3D real-time motion correction hippocampal CA1 pyramidal neurons voltage-calcium coupling in vivo dendritic imaging acousto-optic deflector soma-dendrite attenuation branch-order-dependent signaling Get full text access Log in, subscribe or purchase for full access. References 1. Stuart, G.J. ∙ Spruston, N. Dendritic integration: 60 years of progress Nat. Neurosci. 2015; 18:1713-1721 2. Major, G.

Abstract The distributed architecture of the cortex poses a fundamental challenge for reinforcement learning: how to assign credit specifically to regions that contribute to successful behavior? Cortical neurons can be driven by both global reinforcers, like rewards, and local sensory features, making it difficult to disentangle these influences.